diff --git a/src/learners/gradient_descent.jl b/src/learners/gradient_descent.jl
index be39718..4d6e84f 100644
--- a/src/learners/gradient_descent.jl
+++ b/src/learners/gradient_descent.jl
@@ -125,7 +125,7 @@ replacement for :epochs is specified, or `Δepochs` is negative, instead return
 Return an updated model, with the weights and bias of the previously learned perceptron
 used as the starting state in new gradient descent updates. Adopt any specified
 hyperparameter `replacements` (properties of `LearnAPI.learner(model)`). Exactly `n` new
-epochs are applied, where `n = model.epochs` unless this explictly changed in
+epochs are applied, where `n = model.epochs` unless this explicitly changed in
 `replacements`.
 
 """
diff --git a/src/learners/static_algorithms.jl b/src/learners/static_algorithms.jl
index 8ef9a4d..7c27878 100644
--- a/src/learners/static_algorithms.jl
+++ b/src/learners/static_algorithms.jl
@@ -110,7 +110,7 @@ rejected(model::FancySelectorFitted) = model.rejected
 LearnAPI.fit(learner::FancySelector; verbosity=LearnAPI.default_verbosity()) =
     FancySelectorFitted(learner)
 
-# output the filtered table and add `rejected` field to model (mutatated!)
+# output the filtered table and add `rejected` field to model (mutated!)
 function LearnAPI.transform(model::FancySelectorFitted, X)
     table = Tables.columntable(X)
     names = Tables.columnnames(table)
diff --git a/src/tools.jl b/src/tools.jl
index 910e939..5d4eb7f 100644
--- a/src/tools.jl
+++ b/src/tools.jl
@@ -90,7 +90,7 @@ learner_get(learner, data, apply=identity) =
     get(learner, data, LearnAPI.data_interface(learner), apply)
 
 """
-    model_get(model, data)
+    model_get(model, data; apply=identity)
 
 *Private method.*